Merge branch 'mmagic' into sd-inferencer

Conflicts: tests/test_mmagic.py
open-mmlab · Apr 24, 2023 · c063cb9 · c063cb9
2 parents 2c0aa63 + babe8ff
commit c063cb9
Show file tree

Hide file tree

Showing 170 changed files with 382 additions and 416 deletions.
diff --git a/README.md b/README.md
@@ -189,9 +189,11 @@ python -c "import mmagic; print(mmagic.__version__)"
 ```
 
 **Getting Started**
+
 Please see [quick run](docs/en/get_started/quick_run.md) and [inference](docs/en/user_guides/inference.md) for the basic usage of MMagic.
 
 **Install MMagic from source**
+
 You can also experiment on the latest developed version rather than the stable release by installing MMagic from source with the following commands:
 
 ```shell
@@ -395,7 +397,7 @@ If MMagic is helpful to your research, please cite it as below.
 
 ```bibtex
 @misc{mmagic2023,
-    title = {{MMagic}: {OpenMMLab} Image and Video Editing Toolbox},
+    title = {{MMagic}: {OpenMMLab} Multimodal Advanced, Generative, and Intelligent Creation Toolbox},
     author = {{MMagic Contributors}},
     howpublished = {\url{https://github.com/open-mmlab/mmagic}},
     year = {2023}

diff --git a/README_zh-CN.md b/README_zh-CN.md
@@ -102,7 +102,7 @@
 - 支持本地可视化以及使用 tensorboard 或 wandb的可视化.
 - 支持 33+ 算法 Pytorch 2.0 加速.
 
-**MMagic** 已经支持了[MMEditing](https://github.com/open-mmlab/mmediting) and [MMGeneration](https://github.com/open-mmlab/mmgeneration)和[MMGeneration](https://github.com/open-mmlab/mmgeneration)中的全量任务、模型、优化函数和评价指标 ，并基于[MMEngine](https://github.com/open-mmlab/mmengine)统一了各组件接口 😍。
+**MMagic** 已经支持了[MMEditing](https://github.com/open-mmlab/mmediting)和[MMGeneration](https://github.com/open-mmlab/mmgeneration)中的全量任务、模型、优化函数和评价指标 ，并基于[MMEngine](https://github.com/open-mmlab/mmengine)统一了各组件接口 😍。
 
 如果想了解更多版本更新细节和历史信息，请阅读[更新日志](docs/zh_cn/changelog.md)。如果想从[旧版本](https://github.com/open-mmlab/mmagic/tree/master) MMEditing 0.x 迁移到新版本 MMagic 1.x，请阅读[迁移文档](docs/zh_cn/migration/overview.md)。
 
@@ -190,6 +190,7 @@ python -c "import mmagic; print(mmagic.__version__)"
 请参考[快速运行](docs/zh_cn/get_started/quick_run.md)和[推理演示](docs/zh_cn/user_guides/inference.md)获取 MMagic 的基本用法。
 
 **从源码安装 MMagic**
+
 使用以下命令从源码安装 MMagic，你可以选择不使用已发布的稳定版本，而在最新开发的版本上进行实验。
 
 ```
@@ -391,7 +392,7 @@ MMagic 是一款由不同学校和公司共同贡献的开源项目。我们感
 
 ```bibtex
 @misc{mmagic2023,
-    title = {{MMagic}: {OpenMMLab} Image and Video Editing Toolbox},
+    title = {{MMagic}: {OpenMMLab} Multimodal Advanced, Generative, and Intelligent Creation Toolbox},
     author = {{MMagic Contributors}},
     howpublished = {\url{https://github.com/open-mmlab/mmagic}},
     year = {2023}

diff --git a/configs/_base_/datasets/basicvsr_test_config.py b/configs/_base_/datasets/basicvsr_test_config.py
@@ -5,7 +5,7 @@
     dict(type='GenerateSegmentIndices', interval_list=[1]),
     dict(type='LoadImageFromFile', key='img', channel_order='rgb'),
     dict(type='LoadImageFromFile', key='gt', channel_order='rgb'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 reds_dataloader = dict(
@@ -39,7 +39,7 @@
     dict(type='LoadImageFromFile', key='img', channel_order='rgb'),
     dict(type='LoadImageFromFile', key='gt', channel_order='rgb'),
     dict(type='MirrorSequence', keys=['img']),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 vimeo_90k_bd_dataloader = dict(
@@ -96,7 +96,7 @@
         filename_tmpl='{:04d}.png'),
     dict(type='LoadImageFromFile', key='img', channel_order='rgb'),
     dict(type='LoadImageFromFile', key='gt', channel_order='rgb'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 udm10_dataloader = dict(
@@ -123,7 +123,7 @@
     dict(type='GenerateSegmentIndices', interval_list=[1]),
     dict(type='LoadImageFromFile', key='img', channel_order='rgb'),
     dict(type='LoadImageFromFile', key='gt', channel_order='rgb'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 vid4_bd_dataloader = dict(
     num_workers=1,

diff --git a/configs/_base_/datasets/cifar10_noaug.py b/configs/_base_/datasets/cifar10_noaug.py
@@ -1,4 +1,4 @@
-cifar_pipeline = [dict(type='PackEditInputs')]
+cifar_pipeline = [dict(type='PackInputs')]
 cifar_dataset = dict(
     type='CIFAR10',
     data_root='./data',

diff --git a/configs/_base_/datasets/cifar10_nopad.py b/configs/_base_/datasets/cifar10_nopad.py
@@ -1,6 +1,6 @@
 cifar_pipeline = [
     dict(type='Flip', keys=['gt'], flip_ratio=0.5, direction='horizontal'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 cifar_dataset = dict(
     type='CIFAR10',
@@ -10,7 +10,7 @@
     pipeline=cifar_pipeline)
 
 # test dataset do not use flip
-cifar_pipeline_test = [dict(type='PackEditInputs')]
+cifar_pipeline_test = [dict(type='PackInputs')]
 cifar_dataset_test = dict(
     type='CIFAR10',
     data_root='./data',

diff --git a/configs/_base_/datasets/deblurring-defocus_test_config.py b/configs/_base_/datasets/deblurring-defocus_test_config.py
@@ -23,7 +23,7 @@
         color_type='color',
         channel_order='rgb',
         imdecode_backend='cv2'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 dpdd_data_root = 'data/DPDD'

diff --git a/configs/_base_/datasets/deblurring-motion_test_config.py b/configs/_base_/datasets/deblurring-motion_test_config.py
@@ -11,7 +11,7 @@
         color_type='color',
         channel_order='rgb',
         imdecode_backend='cv2'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 gopro_data_root = 'data/GoPro'

diff --git a/configs/_base_/datasets/decompression_test_config.py b/configs/_base_/datasets/decompression_test_config.py
@@ -17,7 +17,7 @@
         params=dict(quality=[quality, quality]),
         bgr2rgb=True,
         keys=['img']),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 classic5_data_root = 'data/Classic5'

diff --git a/configs/_base_/datasets/denoising-gaussian_color_test_config.py b/configs/_base_/datasets/denoising-gaussian_color_test_config.py
@@ -20,7 +20,7 @@
             gaussian_sigma=[sigma, sigma],
             gaussian_gray_noise_prob=0),
         keys=['img']),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 data_root = 'data/denoising_gaussian_test'

diff --git a/configs/_base_/datasets/denoising-gaussian_gray_test_config.py b/configs/_base_/datasets/denoising-gaussian_gray_test_config.py
@@ -22,7 +22,7 @@
             gaussian_sigma=[sigma, sigma],
             gaussian_gray_noise_prob=1),
         keys=['img']),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 data_root = 'data/denoising_gaussian_test'

diff --git a/configs/_base_/datasets/denoising-real_test_config.py b/configs/_base_/datasets/denoising-real_test_config.py
@@ -11,7 +11,7 @@
         color_type='color',
         channel_order='rgb',
         imdecode_backend='cv2'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 sidd_data_root = 'data/SIDD'

diff --git a/configs/_base_/datasets/deraining_test_config.py b/configs/_base_/datasets/deraining_test_config.py
@@ -11,7 +11,7 @@
         color_type='color',
         channel_order='rgb',
         imdecode_backend='cv2'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 rain100h_data_root = 'data/Rain100H'

diff --git a/configs/_base_/datasets/ffhq_flip.py b/configs/_base_/datasets/ffhq_flip.py
@@ -3,12 +3,12 @@
 train_pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
     dict(type='Flip', keys=['gt'], direction='horizontal'),
-    dict(type='PackEditInputs', keys='gt')
+    dict(type='PackInputs', keys='gt')
 ]
 
 val_pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
-    dict(type='PackEditInputs', keys=['gt'])
+    dict(type='PackInputs', keys=['gt'])
 ]
 
 # `batch_size` and `data_root` need to be set.

diff --git a/configs/_base_/datasets/grow_scale_imgs_ffhq_styleganv1.py b/configs/_base_/datasets/grow_scale_imgs_ffhq_styleganv1.py
@@ -3,7 +3,7 @@
 pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
     dict(type='Flip', keys='gt', direction='horizontal'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 train_dataloader = dict(

diff --git a/configs/_base_/datasets/imagenet_128.py b/configs/_base_/datasets/imagenet_128.py
@@ -8,14 +8,14 @@
     dict(type='RandomCropLongEdge', keys='gt'),
     dict(type='Resize', scale=(128, 128), keys='gt', backend='pillow'),
     dict(type='Flip', keys='gt', flip_ratio=0.5, direction='horizontal'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 test_pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
     dict(type='CenterCropLongEdge', keys='gt'),
     dict(type='Resize', scale=(128, 128), keys='gt', backend='pillow'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 train_dataloader = dict(

diff --git a/configs/_base_/datasets/imagenet_256.py b/configs/_base_/datasets/imagenet_256.py
@@ -8,14 +8,14 @@
     dict(type='RandomCropLongEdge', keys=['img']),
     dict(type='Resize', scale=(256, 256), keys=['img'], backend='pillow'),
     dict(type='Flip', keys=['img'], flip_ratio=0.5, direction='horizontal'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 test_pipeline = [
     dict(type='LoadImageFromFile', key='img'),
     dict(type='CenterCropLongEdge', keys=['img']),
     dict(type='Resize', scale=(256, 256), backend='pillow'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 train_dataloader = dict(

diff --git a/configs/_base_/datasets/imagenet_512.py b/configs/_base_/datasets/imagenet_512.py
@@ -8,14 +8,14 @@
     dict(type='RandomCropLongEdge', keys='gt'),
     dict(type='Resize', scale=(512, 512), keys='gt', backend='pillow'),
     dict(type='Flip', keys='gt', flip_ratio=0.5, direction='horizontal'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 test_pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
     dict(type='CenterCropLongEdge', keys='gt'),
     dict(type='Resize', scale=(512, 512), keys='gt', backend='pillow'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 train_dataloader = dict(

diff --git a/configs/_base_/datasets/imagenet_64.py b/configs/_base_/datasets/imagenet_64.py
@@ -8,14 +8,14 @@
     dict(type='RandomCropLongEdge', keys='gt'),
     dict(type='Resize', scale=(64, 64), keys='gt', backend='pillow'),
     dict(type='Flip', keys='gt', flip_ratio=0.5, direction='horizontal'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 test_pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
     dict(type='CenterCropLongEdge', keys='gt'),
     dict(type='Resize', scale=(64, 64), keys='gt', backend='pillow'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 train_dataloader = dict(

diff --git a/configs/_base_/datasets/imagenet_noaug_128.py b/configs/_base_/datasets/imagenet_noaug_128.py
@@ -9,14 +9,14 @@
     dict(type='LoadImageFromFile', key='img'),
     dict(type='CenterCropLongEdge'),
     dict(type='Resize', scale=(128, 128), backend='pillow'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 test_pipeline = [
     dict(type='LoadImageFromFile', key='img'),
     dict(type='CenterCropLongEdge'),
     dict(type='Resize', scale=(128, 128), backend='pillow'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 train_dataloader = dict(

diff --git a/configs/_base_/datasets/liif_test_config.py b/configs/_base_/datasets/liif_test_config.py
@@ -10,7 +10,7 @@
     dict(
         type='RandomDownSampling', scale_min=scale_test, scale_max=scale_test),
     dict(type='GenerateCoordinateAndCell', scale=scale_test, reshape_gt=False),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ] for scale_test in scale_test_list]
 
 # test config for Set5

diff --git a/configs/_base_/datasets/lsun_stylegan.py b/configs/_base_/datasets/lsun_stylegan.py
@@ -2,12 +2,12 @@
 
 train_pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 val_pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 # `batch_size` and `data_root` need to be set.

diff --git a/configs/_base_/datasets/paired_imgs_256x256_crop.py b/configs/_base_/datasets/paired_imgs_256x256_crop.py
@@ -28,7 +28,7 @@
     #         f'img_{domain_b}': f'img_{domain_b}'
     #     }),
     # dict(
-    #     type='PackEditInputs',
+    #     type='PackInputs',
     #     keys=[f'img_{domain_a}', f'img_{domain_b}'],
     #     data_keys=[f'img_{domain_a}', f'img_{domain_b}'])
 ]
@@ -64,7 +64,7 @@
     #         f'img_{domain_b}': f'img_{domain_b}'
     #     }),
     # dict(
-    #     type='PackEditInputs',
+    #     type='PackInputs',
     #     keys=[f'img_{domain_a}', f'img_{domain_b}'],
     #     data_keys=[f'img_{domain_a}', f'img_{domain_b}'])
 ]

diff --git a/configs/_base_/datasets/sisr_x2_test_config.py b/configs/_base_/datasets/sisr_x2_test_config.py
@@ -11,7 +11,7 @@
         color_type='color',
         channel_order='rgb',
         imdecode_backend='cv2'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 # test config for Set5

diff --git a/configs/_base_/datasets/sisr_x3_test_config.py b/configs/_base_/datasets/sisr_x3_test_config.py
@@ -11,7 +11,7 @@
         color_type='color',
         channel_order='rgb',
         imdecode_backend='cv2'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 # test config for Set5

diff --git a/configs/_base_/datasets/sisr_x4_test_config.py b/configs/_base_/datasets/sisr_x4_test_config.py
@@ -11,7 +11,7 @@
         color_type='color',
         channel_order='rgb',
         imdecode_backend='cv2'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 # test config for Set5

diff --git a/configs/_base_/datasets/tdan_test_config.py b/configs/_base_/datasets/tdan_test_config.py
@@ -5,7 +5,7 @@
     dict(type='GenerateFrameIndiceswithPadding', padding='reflection'),
     dict(type='LoadImageFromFile', key='img', channel_order='rgb'),
     dict(type='LoadImageFromFile', key='gt', channel_order='rgb'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 SPMC_bd_dataloader = dict(
@@ -63,7 +63,7 @@
     dict(type='GenerateFrameIndiceswithPadding', padding='reflection'),
     dict(type='LoadImageFromFile', key='img', channel_order='rgb'),
     dict(type='LoadImageFromFile', key='gt', channel_order='rgb'),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 vid4_bd_dataloader = dict(
     num_workers=1,

diff --git a/configs/_base_/datasets/unconditional_imgs_128x128.py b/configs/_base_/datasets/unconditional_imgs_128x128.py
@@ -4,7 +4,7 @@
 train_pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
     dict(type='Resize', keys='gt', scale=(128, 128)),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 # `batch_size` and `data_root` need to be set.

diff --git a/configs/_base_/datasets/unconditional_imgs_64x64.py b/configs/_base_/datasets/unconditional_imgs_64x64.py
@@ -3,7 +3,7 @@
 train_pipeline = [
     dict(type='LoadImageFromFile', key='gt'),
     dict(type='Resize', keys='gt', scale=(64, 64)),
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 # `batch_size` and `data_root` need to be set.

diff --git a/configs/_base_/datasets/unconditional_imgs_flip_512x512.py b/configs/_base_/datasets/unconditional_imgs_flip_512x512.py
@@ -5,7 +5,7 @@
     dict(type='LoadImageFromFile', key='gt'),
     dict(type='Resize', keys='gt', scale=(512, 512)),
     dict(type='Flip', keys=['gt'], direction='horizontal'),  # TODO:
-    dict(type='PackEditInputs')
+    dict(type='PackInputs')
 ]
 
 # `batch_size` and `data_root` need to be set.